DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 10/10/2025 has been entered.
Claims 1, 5-6, 9, 11, 17 and 19-20 are presently amended.
Claims 1-20 are pending.
Response to Amendment
Applicant’s amendments are acknowledged.
Response to Arguments
Applicant' s arguments filed 10/10/2025 have been fully considered in view of further consideration of statutory law, Office policy, precedential common law, and the cited prior art as necessitated by the amendments to the claims, and are not persuasive for the reasons set forth below.
35 USC § 101 Rejections
First, Applicant argues that, with regard to “…Step 2A Prong 1: Claims Do Not Recite Abstract Ideas… the amended independent claims 1, 11, and 19 are patent-eligible under 35 U.S.C. § 101… The specification clearly describes a significant technical problem in retail shrink prediction technology… These are not merely "identifying and displaying risks for shrinkage in a store," as the Examiner characterized them, but rather sophisticated computer-implemented systems with specific technical implementations…
The amended claims recite "training a machine learning model" generally, similar to eligible Example 39, rather than specifically invoking mathematical concepts like "backpropagation" or "gradient descent" as in ineligible Example 47 claim 2. However, the claims go further by reciting the specific "Random Forest regression model" which provides technological specificity without crossing into abstract mathematical territory…” [Arguments, pages 8-10].
In response, Applicant’s arguments are considered but are not persuasive. Examiner respectfully disagrees and maintains that the present invention recites a judicial exception without significantly more. In particular, Examiner observes that the invention, when considered as a whole, is directed towards “real-time risk assessments” with regard to inventory shrinkage in a physical storefront.
These concepts are not meaningfully different than the following concepts identified by the MPEP: Concepts relating to certain methods of organizing human activity.
The aforementioned claims describe steps for fundamental economic principles or practices, which includes hedging, insurance and mitigating. Specifically, identifying and displaying risks for shrinkage in a store is considered to describe steps for mitigating risk of losses. Examiner further maintains that the machine-learning elements of the claims do not significantly alter the thrust of the invention. Thus, claims 1, 11 and 19 recite concepts identified as abstract ideas. As such, Examiner remains unpersuaded.
Second, Applicant argues that “The amended claims integrate any judicial exception into practical applications through multiple technical means:
Implementing a Particular Machine: The claims recite a specific "Random Forest regression model and one or more data-cleaning algorithms that are based on detection of outliers, wherein dynamic thresholds are applied to determine statistical value of a given feature and set of features to ensure a well-trained MLM" This is not a generic computer implementation but a particular machine tailored for retail shrink prediction with specific algorithmic components.
Improving Computer Technology: The specification teaches that the system improves retail security technology through specific technical implementations. The "MLM provides the time series for a business day at predetermined intervals of time and continuously updates further time intervals for the business day based on current data being received from the store" represents a technological improvement in predictive systems that provides real-time, continuously updating predictions.
Meaningful Technological Application: The claims go beyond merely linking abstract
ideas to technological environments by reciting specific implementations: "assigning a unique color based on the scores to icons or graphical images and superimposing the icons or images onto a background image of the store's layout provided by a planogram, wherein combinations of features are selectable from the interactive interface for a user to visualize a corresponding combination score” [Arguments, page 10].
In response, Applicant’s arguments are considered but are not persuasive. Examiner respectfully disagrees and maintains that the present invention recites a judicial exception without significantly more. In Step 2A Prong Two, Examiners evaluate integration into a practical application by: (1) identifying whether there are any additional elements recited in the claim beyond the judicial exception(s); and (2) evaluating those additional elements individually and in combination to determine whether they integrate the exception into a practical application.
Examiner observes that presently amended claims 1, 11 and 19 only recite the following additional elements –
…training a machine learning model (MLM) comprising a Random Forest regression model…; …a well-trained MLM…; …to the MLM…; …from the MLM…; …icons or graphical images… [Claim 1],
… store systems…; … to a machine-learning model (MLM)…; … from the MLM…; … an interactive interface… icons or graphical images… the icons or images… [Claim 11],
… A system, comprising: a cloud server comprising at least one processor and a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium comprising executable instructions, wherein the executable instructions, when executed by the at least one processor cause the at least one processor to… …training a machine learning model (MLM)… the MLM…; …to the MLM…; …from the MLM…; …icons or graphical images…; …an interactive interface on a device… [Claim 19].
The dependent claims recite the following new additional elements –
… an application programming interface…. [Claims 3 and 20].
With regard to the assertion that the present claims implement a particular machine and improve computer technology, Examiner respectfully disagrees and maintains that the additional elements, when considered in the context of the claims as a whole, are not detailed at a level of specificity that imposes a meaningful limit on the judicial exception. In particular, claim 1 recites the use of generic and unspecified transaction data, customer data, employee data, and security data as input data into a machine learning model and one or more nonspecific data-cleaning algorithms. Examiner respectfully maintains that these limitations are not sufficient to demonstrate that the claim is more than a drafting effort designed to monopolize the judicial exception.
Examiner further maintains that the systems, machine learning model and executable instructions are recited at a high-level of generality (see MPEP § 2106.05(a)), like the following MPEP example:
iii. Gathering and analyzing information using conventional techniques and displaying the result, TLI Communications, 823 F.3d at 612-13, 118 USPQ2d at 1747-48;
Furthermore, the computer implemented element is considered to amount to no more than mere instructions to apply the exception using a generic computer component (see MPEP 2106.05(f)), like the following MPEP example:
i. A commonplace business method or mathematical algorithm being applied on a general purpose computer, Alice Corp. Pty. Ltd. V. CLS Bank Int’l, 573 U.S. 208, 223, 110 USPQ2d 1976, 1983 (2014); Gottschalk v. Benson, 409 U.S. 63, 64, 175 USPQ 673, 674 (1972); Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015).
Accordingly, these additional elements do not integrate the abstract idea into a practical application.
The remaining dependent claims do not recite any new additional elements, and thus do not integrate the abstract idea into a practical application. As such, Examiner remains unpersuaded.
Third, Applicant argues that “The additional elements amount to significantly more than any judicial exception:
Unconventional Machine Learning Implementation: The combination of Random Forest regression with outlier-based data cleaning and dynamic threshold application represents an unconventional approach to retail shrink prediction. This approach is similar to eligible claims in BASCOM, which recited "a specific, discrete implementation" that was "not conventional or generic."
Real-Time Interactive Visualization System: The claims recite specific technical implementations, including "combinations of features are selectable from the interactive interface for a user to visualize a corresponding combination score as the unique color associated with selected features" and continuous time series updates that constitute significantly more than well- understood, routine activity.
Specific Technical Improvements: The claims recite improvements to retail security technology through the specific combination of machine learning algorithms, real-time data processing, and interactive visualization that goes beyond generic computer implementation” [Arguments, pages 10-11].
In response, Applicant’s arguments are considered but are not persuasive. Examiner respectfully disagrees and maintains that the present invention recites a judicial exception without significantly more. In particular and as stated in response to the above argument, claim 1 recites the use of generic and unspecified transaction data, customer data, employee data, and security data as input data into a machine learning model and one or more nonspecific data-cleaning algorithms. Examiner respectfully maintains that these limitations are not sufficient to demonstrate that the claim is more than a drafting effort designed to monopolize the judicial exception. Similarly, Examiner respectfully maintains that the use of real-time store data and color-coded graphics is not sufficient to demonstrate an improvement to any particular technology or to the functioning of computers. Thus, claims 1, 11 and 19 recite judicial exceptions without significantly more. As such, Examiner remains unpersuaded.
35 USC § 103 Rejections
First, Applicant argues that “The amended claim 1 now specifically recites training "a machine learning model (MLM) comprising a Random Forest regression model… While Lobo mentions machine learning generally at 30, it does not disclose Random Forest regression models specifically. Lobo's discussion of machine learning is generic…
The amended claim 1 also recites "assigning a unique color based on the current scores to icons or graphical images representing certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores." … Neither Lobo nor Ozkan teaches unique color assignment to specific icons or graphical images based on shrink risk scores. While Lobo mentions "heat maps" generally at 17, it does not teach the specific visual representation system claimed. Ozkan's color assignment at 137 is based on user traffic density, not shrink risk scores for specific graphical representations of store resources” [Arguments, pages 11-12].
In response, Applicant’s arguments are considered but are not persuasive. First, with respect to the assertion that “Neither Lobo nor Ozkan teaches unique color assignment to specific icons or graphical images based on shrink risk scores”, Examiner respectfully disagrees and maintains that the prior art of record renders the amended claim obvious. In particular, Examiner observes that the color-coded heat map elements of Ozkan (¶¶ 8, 137, 140), in combination with the shrink risk score elements of Lobo (¶¶ 17-19, 25) discloses the above-argued claim elements.
Further, with respect to the random forest elements, Applicant’s arguments have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. As such, Examiner remains unpersuaded.
Second, Applicant argues that “The amended claim 11 recites "assigning a unique color based on the scores to icons or graphical images and superimposing the icons or images onto a background image of a store's layout provided by a planogram…” … Lobo's system appears to provide static reports and alerts, as described in 17: "Embodiments can provide reports, including heat maps, to store managers and associates." This does not suggest the dynamic, interactive interface claimed.
Ozkan's heat maps display traffic density without user-selectable feature combinations. Ozkan at 137 describes assigning colors based on "the number of unique user records," but does not teach interactive selection of feature combinations to visualize combination-specific shrink risk scores overlaid on planograms” [Arguments, page 12].
In response, Applicant’s arguments are considered but are not persuasive. As stated in response to the above argument, Examiner respectfully maintains that the color-coded heat map elements of Ozkan (¶¶ 8, 137, 140), in combination with the shrink risk score elements of Lobo (¶¶ 17-19, 25) renders the above-argued claim elements obvious. Examiner further directs the Applicant to (Ozkan, ¶ 152, The UI application can be executed by the processor 2604 to aid a user in interacting with at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other data associated with the indoor environment, the environment analytics system 118, the network 116, and/or the other systems 122. The UI application can be executed by the processor 2604 to aid a user in answering/initiating calls, entering/deleting other data, entering and setting user IDs and passwords for device access, configuring settings, manipulating address book content and/or settings, multimode interaction, interacting with other applications 2610, and otherwise facilitating user interaction with the operating system 2608, the applications 2610, and/or other types or instances of data 2612 that can be stored at the mobile device 2600. The data 2612 can include, for example, at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other applications or program modules).
Here, Ozkan discloses an interactive user interface to aid a user in interacting with data including the heat map data. Thus, Examiner respectfully maintains that the art of record renders the above-argued claim elements obvious. As such, Examiner remains unpersuaded.
Third, Applicant argues that “The amended claim 19 recites that "the MLM provides the time series for a business day…” … While Lobo mentions predictive modeling at 30, it does not teach time series generation with continuous updates at predetermined intervals throughout a business day. Lobo's system appears to provide alerts when high-risk situations are detected, but it does not describe the specific temporal prediction system with continuous business-day updates as claimed…
Ozkan's system tracks customer movement patterns but does not provide the sophisticated time series prediction with continuous updates for shrink risk assessment throughout business hours, as claimed. ” [Arguments, page 12-13].
In response, Applicant’s arguments are considered but are not persuasive. Examiner respectfully disagrees and directs the Applicant to (Lobo, ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features (discloses customer data and employee data). Real-time sensor data 108 can also include data related to movements of individuals or groups, (discloses security data) congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses transaction data) length of trip data, or other potentially relevant tracked information), (Id., ¶ 31, Referring to FIG. 2, a retail environment 102 is depicted in which a plurality of sensors 104 is located in various locations throughout the retail environment 102. Sensors 104 may be located in fixed locations or may be mobile. In some embodiments, sensors 104 can solely or primarily include security cameras located throughout a store. Alternatively or additionally, sensors 104 can include infrared sensors, optical sensors, temperature sensors, pressure sensors, or other non-intrusive sensors).
Here, Lobo specifies the use of real-time sensor data including time of purchase and length of trip data, for example. Thus, Lobo’s predictions of providing alerts when high-risk situations are detected, while using real-time sensor data, are not considered to be meaningfully different than the above-argued claim limitations. Thus, Examiner respectfully maintains that the art of record renders the above-argued claim elements obvious. As such, Examiner remains unpersuaded.
Fourth, Applicant argues that “The proposed combination lacks proper motivation and would require impermissible hindsight. Lobo focuses on shrink prevention through sensor control and individual risk assessment at 30, while Ozkan addresses customer traffic analysis for marketing optimization at 58… Lobo seeks to prevent theft through sensor control and alerts, while Ozkan seeks to optimize store layout through traffic analysis. No teaching would motivate one skilled in the art to combine Lobo's specific Random Forest regression approach with Ozkan's planogram-based visualization for the specific purpose of interactive shrink risk prediction…” [Arguments, page 13].
In response, Applicant’s arguments are considered but are not persuasive. First, with respect to the assertion that the combination of references would require impermissible hindsight, it must be recognized that any judgment on obviousness is in a sense necessarily a reconstruction based upon hindsight reasoning. But so long as it takes into account only knowledge which was within the level of ordinary skill at the time the claimed invention was made, and does not include knowledge gleaned only from the applicant's disclosure, such a reconstruction is proper. See In re McLaughlin, 443 F.2d 1392, 170 USPQ 209 (CCPA 1971).
Examiner maintains that it would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the heatmap interface elements of Ozkan in the analogous art of determining indoor location of devices using sensors.
The motivation for doing so would have been to “aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results” [Ozkan, ¶ 58], wherein such benefits would benefit Lobo’s method which seeks to improve an “ability to deter shrinkage from stores in a proactive way would be extremely useful to retailers while also improving retail customer experience” [Ozkan, ¶ 58; Lobo, ¶ 4]. As such, Examiner remains unpersuaded.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Step 1: Claims 1-20 are directed to statutory categories, namely processes (claims 1-10 and 11-18), and a machine (claims 19-20).
Step 2A, Prong 1: Claims 1, 11 and 19 in part, recite the following abstract idea:
…A method, comprising: obtaining, as input, transaction data, customer data, employee data, and security data as input data for a store; …and one or more data-cleaning algorithms that are based on detection of outliers, wherein dynamic thresholds are applied to determine statistical value of a given feature and set of features to ensure … for detecting dimensions or features of shrink on features derived from the input data and known shrink events identified in the security data to produce, as output, shrink scores for the features and for combinations of the features; obtaining real-time transaction data and security data from the store as current input data; extracting current features from the current input data; providing the current features as input …; receiving current scores as output …; generating a data structure depicting a physical layout of the store with the features uniquely visually identified based on their corresponding current scores and overlaid on the physical layout; rendering the data structure within an interface accessible by a store user; and assigning a unique color based on the current scores to icons or graphical images representing to certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores. [Claim 1],
A method, comprising: deriving features indicative of shrink from store data that is obtained from … and from data stores of a store; providing the features as input… ; receiving as output … scores that predict whether shrink is likely to occur or not for each of the features and for combinations of the features within a given interval of future time beginning with a current time; generating a heatmap depicting resources associated with the features and resource locations for the resources within a physical layout of the store; assigning indicia to each of the resources based on a corresponding score assigned to a corresponding feature; providing a function associated with the heatmap, wherein the function when accessed from the heatmap performs operations comprising: animating from the current time to an end of the given interval of future time; and visually depicting risk associated with shrink within a context of the physical layout of the store through changing colors associated with the resources over time based on the animating; rendering the heatmap with the function within … to a store user; and assigning a unique color based on the scores to … and superimposing … onto a background image of a store's layout provided by a planogram, wherein combinations of features are selectable from the interactive interface for a user to visualize a corresponding combination score as the unique color associated with selected features based on a corresponding score assigned to the corresponding feature [Claim 11],
…perform operations comprising: … on features derived from store data obtained from store systems and data stores of a store to produce as output a time series of sets of scores over a predefined period of time that predicts at each interval a corresponding set of scores indicating how likely each feature or combinations of the features is likely or not likely to be associated with shrink during a corresponding interval wherein … provides the time series for a business day at predetermined intervals of time and continuously updates further time intervals for the business day based on current data being received from the store; receiving real-time store data at a current interval of time from the store systems and the data stores; deriving current features from the real-time store data; providing the current features as input …; receiving as output… current sets of current scores for current features and current combinations of the features, wherein a last set of the current sets of current scores ends at a close of business hours for the store; obtaining a planogram for a physical layout of the store with resource identifiers for resources of the store identified within the physical layout; generating a heatmap data structure with labels or images for the resource identifiers superimposed on top of the physical layout; assigning a unique color based on the current sets of current scores to …representing certain resources and certain combinations of the resources that are associated with the current features and the current combinations based on corresponding current scores; providing a function with the heatmap data structure to animate the certain resources, the certain combinations, and corresponding indicia within the physical layout over time until the close of the business hours for the store; rendering the heatmap data structure within … operated by a store user for the store user to visually identify areas within the store, the certain resources, and combinations of the certain resources that are associated with high risk of shrink throughout the business hours of the store; and obtaining external data associated with dates of transactions and a physical location of the store as second features. [Claim 19].
These concepts are not meaningfully different than the following concepts identified by the MPEP:
Concepts relating to certain methods of organizing human activity. The aforementioned limitations describe steps for fundamental economic principles or practices, which includes hedging, insurance and mitigating. Specifically, identifying and displaying risks for shrinkage in a store is considered to describe steps for mitigating risk of losses. As such, claims 1, 11 and 19 recite concepts identified as abstract ideas.
The dependent claims recite limitations relative to the independent claims, including, for example:
… iterating to the obtaining of the real-time transaction data and the security data at preconfigured intervals of time [Claim 2],
…sending feature identifiers for the features and the corresponding current scores to a system or an application of the store using … [Claim 3],
…wherein obtaining the transaction data further includes merging and associating the input data by transaction [Claim 4],
…wherein merging further includes labeling each transaction within the input data with any shrink event data associated with a shrink event for a corresponding transaction [Claim 5],
…wherein training further includes providing the shrink event that is labeled as expected predicted output produced by … on the features [Claim 6].
The limitations of these dependent claims are merely narrowing the abstract idea identified in the independent claims, and thus, the dependent claims also recite abstract ideas.
Step 2A, Prong 2: This judicial exception is not integrated into a practical application. In particular, claims 1, 11 and 19 only recite the following additional elements –
…training a machine learning model (MLM) comprising a Random Forest regression model…; …a well-trained MLM…; …to the MLM…; …from the MLM…; …icons or graphical images… [Claim 1],
… store systems…; … to a machine-learning model (MLM)…; … from the MLM…; … an interactive interface… icons or graphical images… the icons or images… [Claim 11],
… A system, comprising: a cloud server comprising at least one processor and a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium comprising executable instructions, wherein the executable instructions, when executed by the at least one processor cause the at least one processor to… …training a machine learning model (MLM)… the MLM…; …to the MLM…; …from the MLM…; …icons or graphical images…; …an interactive interface on a device… [Claim 19].
The dependent claims recite the following new additional elements –
… an application programming interface…. [Claims 3 and 20].
The systems, machine learning model and executable instructions are recited at a high-level of generality (see MPEP § 2106.05(a)), like the following MPEP example:
iii. Gathering and analyzing information using conventional techniques and displaying the result, TLI Communications, 823 F.3d at 612-13, 118 USPQ2d at 1747-48;
Furthermore, the computer implemented element is considered to amount to no more than mere instructions to apply the exception using a generic computer component (see MPEP 2106.05(f)), like the following MPEP example:
i. A commonplace business method or mathematical algorithm being applied on a general purpose computer, Alice Corp. Pty. Ltd. V. CLS Bank Int’l, 573 U.S. 208, 223, 110 USPQ2d 1976, 1983 (2014); Gottschalk v. Benson, 409 U.S. 63, 64, 175 USPQ 673, 674 (1972); Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015).
Accordingly, these additional elements do not integrate the abstract idea into a practical application.
The remaining dependent claims do not recite any new additional elements, and thus do not integrate the abstract idea into a practical application.
Step 2B: Claims 1, 11 and 19 and their underlying limitations, steps, features and terms, considered both individually and as a whole, do not include additional elements that are sufficient to amount to significantly more than the judicial exception for the following reasons:
Independent claims 1, 11 and 19 only recite the following additional elements –
…training a machine learning model (MLM)…; …to the MLM…; …from the MLM…;
… store systems…; … to a machine-learning model (MLM)…; … from the MLM…; … an interactive interface… [Claim 11],
… A system, comprising: a cloud server comprising at least one processor and a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium comprising executable instructions, wherein the executable instructions, when executed by the at least one processor cause the at least one processor to… …training a machine learning model (MLM)…; …to the MLM…; …from the MLM…; …an interactive interface on a device… [Claim 19].
These elements do not amount to significantly more than the abstract idea for the reasons discussed in 2A prong 2 with regard to MPEP 2106.05(a) and MPEP 2106.05(f). By the failure of the elements to integrate the abstract idea into a practical application there, the additional elements likewise fail to amount to an inventive concept that is significantly more than an abstract idea here, in Step 2B.
As such, both individually or in combination, these limitations do not add significantly more to the judicial exception.
The remaining dependent claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the dependent claims do not recite any new additional elements other than those mentioned in the independent claims, which amount to no more than mere instructions to apply the exception using a generic computer component (see MPEP 2106.05(f)). As such, these claims are not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-6 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Lobo et al., U.S. Publication No. 2019/0027003 [hereinafter Lobo], in view of Ozkan et al., U.S. Publication No. 2015/0289111 [hereinafter Ozkan] and in further view of Wong et al., U.S. Publication No. 2023/0118240 [hereinafter Wong].
Regarding Claim 1, Lobo discloses … A method, comprising: obtaining, as input, transaction data, customer data, employee data, and security data as input data for a store (Lobo, ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features (discloses customer data and employee data). Real-time sensor data 108 can also include data related to movements of individuals or groups, (discloses security data) congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses transaction data) length of trip data, or other potentially relevant tracked information), (Id., ¶ 31, Referring to FIG. 2, a retail environment 102 is depicted in which a plurality of sensors 104 is located in various locations throughout the retail environment 102. Sensors 104 may be located in fixed locations or may be mobile. In some embodiments, sensors 104 can solely or primarily include security cameras located throughout a store. Alternatively or additionally, sensors 104 can include infrared sensors, optical sensors, temperature sensors, pressure sensors, or other non-intrusive sensors), (Id., ¶ 39, FIG. 4 shows a diagram of an example of a store network 400 in which a retail shrinkage activity prediction and identification system can be implemented. In FIG. 4, the store network generally includes a store 410 having a main retail environment 412 and a back office 414. The main retail environment 412 includes a rotating system 416 having a plurality of related POS systems 418, cameras 420, and connectivity devices 422. Also connected to the main retail environment are maintenance resources 424);
PNG
media_image1.png
355
355
media_image1.png
Greyscale
training a machine learning model (MLM) … on features derived from the input data and known shrink events identified in the security data to produce, as output, shrink scores for the features and for combinations of the features (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. (discloses machine learning model for retail shrinkage scoring) Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates), (Id., ¶ 18, Some embodiments can turn cameras toward more susceptible items or areas at certain times. Some embodiments can update an internal shrink database to better track and proactively identify high risk items and areas of a store. Some embodiments can also utilize machine learning to identify an individual in a retail store by capturing an image of the individual and searching external or public reports and databases of high-risk individuals or groups, such as those shared among retailers or business associations in particular industries and/or geographic areas or available from government or law enforcement agencies), (Id., ¶ 19, References to “shrinkage” or “shrink,” as used throughout this disclosure, are intended to refer generally to loss of inventory that can be attributed to factors such as theft, shoplifting, administrative errors, fraud, and cashier errors that benefit the purchaser), (Id., ¶ 25, First shrinkage database 120 includes retail shrinkage data 122 for the retail environment 102. The retail shrinkage data 122 can include one or more items 124 at high risk for shrinkage. These may be items that have a history of being stolen frequently, (discloses known shrink events) are particularly valuable, or are known to be related to frequent shrinkage-related issues);
obtaining real-time transaction data and security data from the store as current input data (Id. ¶ 31, Real-time sensor data 108 can include image data, such as an image of clothing or facial features. Real-time sensor data 108 can also include data related to movements of individuals or groups, congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses real-time transaction data) length of trip data, or other potentially relevant tracked information);
extracting current features from the current input data; providing the current features as input to the MLM; receiving current scores as output from the MLM (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur);
generating a data structure depicting a physical layout of the store with the features uniquely visually identified based on their corresponding current scores and overlaid on the physical layout (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates (discloses heatmap data structure));
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose … comprising a Random Forest regression model and one or more data-cleaning algorithms that are based on detection of outliers, wherein dynamic thresholds are applied to determine statistical value of a given feature and set of features to ensure a well-trained MLM for detecting dimensions or features of shrink…; rendering the data structure within an interface accessible by a store user; and assigning a unique color based on the current scores to icons or graphical images representing certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores
However, Ozkan discloses …rendering the data structure within an interface accessible by a store user (Ozkan, ¶ 34, FIG. 21 is a user interface diagram illustrating an example heat map on an example store layout, according to an illustrative embodiment (discloses interface display of heatmap data structure)), (Id., ¶ 150, As illustrated in FIG. 26, the mobile device 2600 can include a display 2602 for displaying data. According to various embodiments, the display 2602 can be configured to display at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, various graphical user interface (“GUI”) elements such as the elements illustrated and described herein with reference to FIG. 12, text, images, video, virtual keypads and/or keyboards, messaging data, notification messages, metadata, internet content, device status, time, date, calendar data, device preferences, map and location data, combinations thereof, and/or the like. The mobile device 2600 also can include a processor 2604 and a memory or other data storage device (“memory”) 2606. The processor 2604 can be configured to process data and/or can execute computer-executable instructions stored in the memory 2606. The computer-executable instructions executed by the processor 2604 can include, for example, an operating system 2608, one or more applications 2610 such as the application 128, other computer-executable instructions stored in a memory 2608, or the like. In some embodiments, the applications 2606 also can include a UI application (not illustrated in FIG. 26)), (Id., ¶ 152, The UI application can be executed by the processor 2604 to aid a user in interacting with at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other data associated with the indoor environment, the environment analytics system 118, the network 116, and/or the other systems 122. The UI application can be executed by the processor 2604 to aid a user in answering/initiating calls, entering/deleting other data, entering and setting user IDs and passwords for device access, configuring settings, manipulating address book content and/or settings, multimode interaction, interacting with other applications 2610, and otherwise facilitating user interaction with the operating system 2608, the applications 2610, and/or other types or instances of data 2612 that can be stored at the mobile device 2600. The data 2612 can include, for example, at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other applications or program modules);
PNG
media_image2.png
431
579
media_image2.png
Greyscale
and assigning a unique color based on the current scores to icons or graphical images representing certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores (Id., ¶ 8, the environment analytics system also can apply a coordinate system to the layout of the environment. The coordinate system can include the absolute reference point. The environment analytics system also can determine a minimum coordinate pair, a maximum coordinate pair, a granularity, and a time interval. The environment analytics system also can set a first coordinate equal to a first minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a second coordinate equal to a second minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a third coordinate equal to a sum of the first minimum coordinate and the granularity, and can set a fourth coordinate equal to a sum of the second minimum coordinate and the granularity. The environment analytics system also can query the environment database for a number of unique user location records with a first location coordinate between the first coordinate and the third coordinate, a second location coordinate between the second coordinate and the fourth coordinate, and a timestamp within the time interval. The environment analytics system also can determine heat map color codes for a plurality of different numbers of unique user location records. The environment analytics system also can generate a heat map that includes a plurality of areas representing at least a portion of the heat map color codes), (Id., ¶ 137, From operation 1912, the method 1900 proceeds to operation 1914, where the environment analytics system 118 assigns a heat map color for each area based upon the number of unique user records. For example, if the number of unique users is 0, the heat map color may be white; for 1-10 unique users, the heat map color may be light yellow; for 11-50 users, the heat map color may be dark yellow; for 51-100 unique users, the heat map color may be orange; for 101-200 unique users, the heat map color may be brown; and for greater than 200 unique users, the heat map color may be red. (discloses assigning colors to an image based on current scores) These colors are provided as examples only and the actual number of unique users for each heat map color may be specified differently for different implementations of the concepts and technologies disclosed herein. An example heat map is shown in grayscale in FIG. 21), (Id., ¶ 140, Turning now to FIG. 22, a method 2200 for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map will be described, according to an illustrative embodiment. The method 2200 begins and proceeds to operation 2202, where the environment analytics system 118 monitors and captures location updates for a plurality of users navigating the indoor environment 102. From operation 2202, the method 2200 proceeds to operation 2204, where the environment analytics system 118 presents the location updates on a display in accordance with color codes associated with each granular area of the indoor environment 102. From operation 2204, the method 2000 proceeds to operation 2206, where the method 2200 ends. The method 2200 shows the location updates on the layout).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the heatmap interface elements of Ozkan in the analogous art of determining indoor location of devices using sensors.
The motivation for doing so would have been to “aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results” [Ozkan, ¶ 58], wherein such benefits would benefit Lobo’s method which seeks to improve an “ability to deter shrinkage from stores in a proactive way would be extremely useful to retailers while also improving retail customer experience” [Ozkan, ¶ 58; Lobo, ¶ 4].
While suggested in at least Fig. 2 and related text of Lobo, the combination of Lobo and Ozkan does not explicitly disclose … comprising a Random Forest regression model and one or more data-cleaning algorithms that are based on detection of outliers, wherein dynamic thresholds are applied to determine statistical value of a given feature and set of features to ensure a well-trained MLM for detecting dimensions or features of shrink.
However, Wong discloses … comprising a Random Forest regression model and one or more data-cleaning algorithms that are based on detection of outliers, wherein dynamic thresholds are applied to determine statistical value of a given feature and set of features to ensure a well-trained MLM for detecting dimensions or features of shrink (Wong, ¶ 49, FIG. 2A shows a machine learning system 210 communicatively coupled to a data bus 220. The data bus 220 may comprise an internal data bus of the machine learning server 150 or may form part of storage area network. The data bus 220 communicatively couples the machine learning system 210 to a plurality of data storage devices 230, 232. The data storage devices 230, 232 may comprise any known data storage device such as magnetic hard disks and solid-state devices. Although data storage devices 230, 232 are shown as different devices in FIG. 2A they may alternatively form different physical areas or portions of storage within a common data storage device. In FIG. 2A, the plurality of data storage devices 230, 232 store historical transaction data 240 and ancillary data 242. In FIG. 2A, a first set of data storage devices 230 store historical transaction data 240 and a second set of data storage devices 232 store ancillary data 242. Ancillary data 242 may comprise one or more of model parameters for a set of machine learning models (such as trained parameters for a neural network architecture and/or configuration parameters for a random forest model) and state data for those models. In one case, the different sets of historical transaction data 240-A to N and ancillary data 242-A to N are associated with different entities that are securely and collectively use services provided by the machine learning system 210, e.g. these may represent data for different banks that need to be kept separate as part of the conditions of providing machine learning services to those entities), (Id., ¶ 103, The second configuration 700 of FIG. 7 is based on a random forest model. The random forest model is applied to an input feature vector 710 and comprises a plurality of decision trees 720 that are applied in parallel. Each decision tree 722, 724, 726 outputs a different classification value Ci 730. Three decision trees 722, 724, and 726 and three classification values 732, 734, and 736 are shown in FIG. 7 but there may be N decision trees in total, where N is a configuration parameter and may number in the hundreds. The classification values 730 are passed to an ensemble processor 740 that combines the classification values 730 from each of the decision trees 720 to generate a final scalar output 750. The ensemble processor 740 may compute a weighted output of the decisions of each decision tree 720 and/or may apply a voting procedure), (Id., ¶ 144, In unsupervised outlier detection, features are generated that are deemed to be informative to quantify deviations from expected behaviour, where the features are input to an anomaly detection system, which is configured to identify features that, within a population of features, are outliers relative to the overall data distribution of those features. Approaches for performing unsupervised outlier detection include using tree-based isolation forests, generative adversarial networks or variational autoencoders. The paper “Variational Autoencoder based Anomaly Detection using Reconstruction Probability” by An and Cho (SNU Data Mining Center—2015-2 Special Lecture on IE), which is incorporated herein by reference, describes an anomaly detection method using the reconstruction probability from the variational autoencoder. However, these approaches are complex and difficult to reconcile with the constraints of transaction processing).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo and the heatmap interface elements of Ozkan in the analogous art of determining indoor location of devices using sensors to include the random forest elements of Wong in the analogous art of transaction data processing.
The motivation for doing so would have been to improve customer service and experience by “ preventing fraud in cases where the physical presence of a payment card cannot be ascertained (e.g., online transactions referred to as “card-not-present”) or for commercial transactions where high-value transactions may be routine and where it may be difficult to classify patterns of behaviour as “unexpected”” [Wong, ¶ 25], wherein such improvements would beneifit Ozkan’s method which seeks to “aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results” [Ozkan, ¶ 58], and wherein such improvements would further benefit Lobo’s method which seeks to improve an “ability to deter shrinkage from stores in a proactive way would be extremely useful to retailers while also improving retail customer experience” [Wong, ¶ 25; Ozkan, ¶ 58; Lobo, ¶ 4].
Regarding Claim 2, the combination of Lobo, Ozkan and Wong discloses …The method of claim 1…
Lobo further discloses …further comprising, iterating to the obtaining of the real-time transaction data and the security data at preconfigured intervals of time (Id., ¶ 46, one or more of the embodiments include one or more localized Internet of Things (IoT) devices and controllers. As a result, in an embodiment, the localized IoT devices and controllers can perform most, if not all, of the computational load and associated monitoring and then later asynchronous uploading of summary data can be performed by a designated one of the IoT devices to a remote server. In this manner, the computational effort of the overall system may be reduced significantly. For example, whenever a localized monitoring device allows remote transmission, secondary utilization of controllers secures data for other IoT devices and permits periodic asynchronous uploading of the summary data to the remote server. In addition, in an exemplary embodiment, the periodic asynchronous uploading of summary data may include a key kernel index summary of the data as created under nominal conditions. In an embodiment, the kernel encodes relatively recently acquired intermittent data (“KRI”). As a result, in an embodiment, KRI includes a source of substantially all continuously-utilized near term data. However, KRI may be discarded depending upon the degree to which such KRI has any value based on local processing and evaluation of such KRI. In an exemplary embodiment, KRI may not even be utilized in any form if it is determined that KRI is transient and may be considered as signal noise), (Id., ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features (discloses customer data and employee data). Real-time sensor data 108 can also include data related to movements of individuals or groups, (discloses security data) congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses transaction data) length of trip data, or other potentially relevant tracked information).
Regarding Claim 3, the combination of Lobo, Ozkan and Wong discloses …The method of claim 1…
Lobo further discloses …further comprising, sending feature identifiers for the features and the corresponding current scores to a system or an application of the store… (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur);
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose … using an application programming interface.
However, Ozkan discloses … using an application programming interface (Ozkan, ¶ 151, The UI application can interface with the operating system 2608 to facilitate user interaction with functionality and/or data stored at the mobile device 2600 and/or stored elsewhere, such as in the environment database 120. In some embodiments, the operating system 2608 can include a member of the SYMBIAN OS family of operating systems from SYMBIAN LIMITED, a member of the WINDOWS MOBILE OS and/or WINDOWS PHONE OS families of operating systems from MICROSOFT CORPORATION, a member of the PALM WEBOS family of operating systems from HEWLETT PACKARD CORPORATION, a member of the BLACKBERRY OS family of operating systems from RESEARCH IN MOTION LIMITED, a member of the IOS family of operating systems from APPLE INC., a member of the ANDROID OS family of operating systems from GOOGLE INC., and/or other operating systems. These operating systems are merely illustrative of some contemplated operating systems that may be used in accordance with various embodiments of the concepts and technologies described herein and therefore should not be construed as being limiting in any way).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the application interface elements of Ozkan in the analogous art of determining indoor location of devices using sensors for the same reasons as stated for claim 1.
Regarding Claim 4, the combination of Lobo, Ozkan and Wong discloses …The method of claim 1…
Lobo further discloses … wherein obtaining the transaction data further includes merging and associating the input data by transaction (Id., ¶ 20, Retail stores or environments in which these shrink mitigation systems and methods can be used include virtually any retail outlet, including a physical, brick-and-mortar storefront; or some other setting or location via which a customer may purchase or obtain products. Though only the case of a single retail environment is generally discussed in examples used herein, in many cases, the systems and methods can include a plurality of retail environments. For example, data from one or a plurality of retail environments can be aggregated, analyzed and applied to one or a plurality of other retail environments. In some embodiments, data from one or a plurality of retail environments can be aggregated, analyzed and/or applied in conjunction with data related to other shopping behaviors, patterns or other factors), (Id., ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features (discloses customer data and employee data). Real-time sensor data 108 can also include data related to movements of individuals or groups, (discloses security data) congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses transaction data) length of trip data, or other potentially relevant tracked information).
Regarding Claim 5, the combination of Lobo, Ozkan and Wong discloses …The method of claim 4…
Lobo further discloses … wherein merging further includes labeling each transaction within the input data with any shrink event data associated with a shrink event for a corresponding transaction (Id., ¶ 20, Retail stores or environments in which these shrink mitigation systems and methods can be used include virtually any retail outlet, including a physical, brick-and-mortar storefront; or some other setting or location via which a customer may purchase or obtain products. Though only the case of a single retail environment is generally discussed in examples used herein, in many cases, the systems and methods can include a plurality of retail environments. For example, data from one or a plurality of retail environments can be aggregated, analyzed and applied to one or a plurality of other retail environments. In some embodiments, data from one or a plurality of retail environments can be aggregated, analyzed and/or applied in conjunction with data related to other shopping behaviors, patterns or other factors), (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates), (Id., ¶ 18, Some embodiments can turn cameras toward more susceptible items or areas at certain times. Some embodiments can update an internal shrink database to better track and proactively identify high risk items and areas of a store. Some embodiments can also utilize machine learning to identify an individual in a retail store by capturing an image of the individual and searching external or public reports and databases of high-risk individuals or groups, such as those shared among retailers or business associations in particular industries and/or geographic areas or available from government or law enforcement agencies).
Regarding Claim 6, the combination of Lobo, Ozkan and Wong discloses …The method of claim 5…
Lobo further discloses … wherein training further includes providing the shrink event that is labeled as expected predicted output produced by the machine learning model based on the features (Id., ¶ 42, Initially at 501, an individual is detected and an image is captured at 503. Next, a procedure is used to identify one or more features of the individual at 505 using the analytics engine. The system is queried as to whether a match for the feature(s) was found at 507. If no match is found at 507, the device control engine tunes the sensors at 509 and tunes the camera at 511. If a match for the feature(s) is found at 507, and a maximum number of attempts for searching the matched feature(s) is not exceeded at 513, the central cloud 540 searches criminal reports and one or more known high risk individual databases at 515. The shrink risk is assessed at 517, and if a high risk of shrink is found, alerts 519 are given. Further, cameras are rotated appropriately to the individual detected at 521 and video is captured of the individual and his or her movements at 523. Further if alerts are given at 519, clients are notified at 525 and criminal reports and known high risk individual databases are updated at 527. This information is fed into the machine learning engine 550 for training at 529 and predictions are made by the machine learning engine 550 at 531), (Id., ¶ 38, At 318, predictive modeling is conducted using the retail shrinkage data 122, the external data 132, and the issuance of an alert 142. Conducting predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. At 320, an alert is issued if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some cases, if the predictive modeling determines that a high shrinkage risk situation is likely to occur, the sensor control system 110 can be caused to alter a setting 106 of at least one of the plurality of sensors 104).
Regarding Claim 10, the combination of Lobo, Ozkan and Wong discloses …The method of claim 1…
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose … wherein generating further includes generating the data structure as a heatmap with a timeline function that permits the heatmap to be animated from a current time to a future time.
However, Ozkan discloses … wherein generating further includes generating the data structure as a heatmap with a timeline function that permits the heatmap to be animated from a current time to a future time (Ozkan, ¶ 34, Fig. 22 is a flow diagram illustrating a method for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map, (discloses animated heatmap) according to an illustrative embodiment), (Id., ¶ 49, The environment analytics system 118 can perform operations to interact with the environment database 120, and more particularly, environment data 132, customer data 134, product data 136, promotion data 138, beacon data 140, heat map data 142 and/or calibration data 144 stored within the environment database 120. The environment analytics system 118 can save data to the environment database 120, retrieve data from the environment database 120, delete data from the environment database 120, edit data and saved edited data to the environment database 120, and manipulate data stored within the environment database 120. The environment analytics system 118 can provide data retrieved from the environment database 120 to the user device 110 and/or the other systems 122 via the network 116), (Id., ¶ 58, The heat map data 142 can include data associated with customer heat maps. Customer heat maps are used herein to identify hot spots, dead areas and bottlenecks of customer traffic within the indoor environment 102. Customer heat maps can aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results. Companies can use heat maps to visualize the impact of changes to the indoor environment in terms of customer flows, sold items, average sales values, and the like. The heat map data 142 can include previously recorded customer coordinate during a given time interval and/or real-time customer coordinate information. (discloses timeline function) It should be understood that the heat map data 142 can include any combination of the aforementioned data and other data associated with heat maps that is not specified herein).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the animated heatmap elements of Ozkan in the analogous art of determining indoor location of devices using sensors for the same reasons as stated for claim 1.
Claims 11-20 are rejected under 35 U.S.C. 103 as being unpatentable over Lobo in view of Ozkan
Regarding Claim 11, Lobo discloses … A method, comprising: deriving features indicative of shrink from store data that is obtained from store systems and from data stores of a store (Lobo, ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates), (Id., ¶ 25, First shrinkage database 120 includes retail shrinkage data 122 for the retail environment 102. The retail shrinkage data 122 can include one or more items 124 at high risk for shrinkage. These may be items that have a history of being stolen frequently, are particularly valuable, or are known to be related to frequent shrinkage-related issues. Items that are small, easy to conceal, or difficult to track could also be deemed items 124 at high risk for shrinkage. The retail shrinkage data 122 can include one or more times 126 at high risk for shrinkage activity. These times 126 can include times of day when shrinkage is most common, times of the week common for shrinkage, times of the year common for shrinkage, or times of expected shrinkage related to holidays and local activities. Further, certain items can be correlated to certain times to identify high shrinkage risk. In some embodiments, at least one item 124 at high risk or at least one time 126 at high risk for shrinkage is part of the retail shrinkage data 122.
providing the features as input to a machine-learning model (MLM); receiving as output from the MLM scores that predict whether shrink is likely to occur or not for each of the features and for combinations of the features within a given interval of future time beginning with a current time (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur);
generating a heatmap depicting resources associated with the features and resource locations for the resources within a physical layout of the store (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates (discloses heatmap data structure)).
assigning indicia to each of the resources based on a corresponding score assigned to the corresponding feature (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur), (Id., ¶ 7, if a medium shrinkage risk situation is identified, the analytics engine will classify it as such and update at least one of the first shrinkage database or the second shrinkage database. In some embodiments, if an existing item in one of the first shrinkage database or the second shrinkage database that was previously identified as high risk or medium risk is identified as low risk, it will be reclassified as low risk (discloses assigning risk indicia based on risk scores) and updated in the first shrinkage database and the second shrinkage database).
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose …providing a function associated with the heatmap, wherein the function when accessed from the heatmap performs operations comprising: animating from the current time to an end of the given interval of future time; and visually depicting risk associated with shrink within a context of the physical layout of the store through changing colors associated with the resources over time based on the animating; rendering the heatmap with the function within an interactive interface to a store user; and assigning a unique color based on the scores to icons or graphical images and superimposing the icons or images onto a background image of a store's layout provided by a planogram, wherein combinations of features are selectable from the interactive interface for a user to visualize a corresponding combination score as the unique color associated with selected features based on a corresponding score assigned to the corresponding feature.
However, Ozkan discloses …providing a function associated with the heatmap, wherein the function when accessed from the heatmap performs operations comprising: animating from the current time to an end of the given interval of future time; and visually depicting risk associated with shrink within a context of the physical layout of the store through changing colors associated with the resources over time based on the animating; rendering the heatmap with the function within an interactive interface to a store user (Ozkan, ¶ 34, Fig. 22 is a flow diagram illustrating a method for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map, (discloses animated heatmap) according to an illustrative embodiment), (Id., ¶ 49, The environment analytics system 118 can perform operations to interact with the environment database 120, and more particularly, environment data 132, customer data 134, product data 136, promotion data 138, beacon data 140, heat map data 142 and/or calibration data 144 stored within the environment database 120. The environment analytics system 118 can save data to the environment database 120, retrieve data from the environment database 120, delete data from the environment database 120, edit data and saved edited data to the environment database 120, and manipulate data stored within the environment database 120. The environment analytics system 118 can provide data retrieved from the environment database 120 to the user device 110 and/or the other systems 122 via the network 116), (Id., ¶ 58, The heat map data 142 can include data associated with customer heat maps. Customer heat maps are used herein to identify hot spots, dead areas and bottlenecks of customer traffic within the indoor environment 102. Customer heat maps can aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results. Companies can use heat maps to visualize the impact of changes to the indoor environment in terms of customer flows, sold items, average sales values, and the like. The heat map data 142 can include previously recorded customer coordinate during a given time interval and/or real-time customer coordinate information. (discloses timeline function) It should be understood that the heat map data 142 can include any combination of the aforementioned data and other data associated with heat maps that is not specified herein), (Id., ¶ 137, From operation 1912, the method 1900 proceeds to operation 1914, where the environment analytics system 118 assigns a heat map color (discloses heat map color coding) for each area based upon the number of unique user records. For example, if the number of unique users is 0, the heat map color may be white; for 1-10 unique users, the heat map color may be light yellow; for 11-50 users, the heat map color may be dark yellow; for 51-100 unique users, the heat map color may be orange; for 101-200 unique users, the heat map color may be brown; and for greater than 200 unique users, the heat map color may be red. These colors are provided as examples only and the actual number of unique users for each heat map color may be specified differently for different implementations of the concepts and technologies disclosed herein. An example heat map is shown in grayscale in FIG. 21);
and assigning a unique color based on the scores to icons or graphical images and superimposing the icons or images onto a background image of a store's layout provided by a planogram, wherein combinations of features are selectable from the interactive interface for a user to visualize a corresponding combination score as the unique color associated with selected features based on a corresponding score assigned to the corresponding feature (Id., ¶ 8, the environment analytics system also can apply a coordinate system to the layout of the environment. The coordinate system can include the absolute reference point. The environment analytics system also can determine a minimum coordinate pair, a maximum coordinate pair, a granularity, and a time interval. The environment analytics system also can set a first coordinate equal to a first minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a second coordinate equal to a second minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a third coordinate equal to a sum of the first minimum coordinate and the granularity, and can set a fourth coordinate equal to a sum of the second minimum coordinate and the granularity. The environment analytics system also can query the environment database for a number of unique user location records with a first location coordinate between the first coordinate and the third coordinate, a second location coordinate between the second coordinate and the fourth coordinate, and a timestamp within the time interval. The environment analytics system also can determine heat map color codes for a plurality of different numbers of unique user location records. The environment analytics system also can generate a heat map that includes a plurality of areas representing at least a portion of the heat map color codes), (Id., ¶ 137, From operation 1912, the method 1900 proceeds to operation 1914, where the environment analytics system 118 assigns a heat map color for each area based upon the number of unique user records. For example, if the number of unique users is 0, the heat map color may be white; for 1-10 unique users, the heat map color may be light yellow; for 11-50 users, the heat map color may be dark yellow; for 51-100 unique users, the heat map color may be orange; for 101-200 unique users, the heat map color may be brown; and for greater than 200 unique users, the heat map color may be red. These colors are provided as examples only and the actual number of unique users for each heat map color may be specified differently for different implementations of the concepts and technologies disclosed herein. An example heat map is shown in grayscale in FIG. 21), (Id., ¶ 140, Turning now to FIG. 22, a method 2200 for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map will be described, according to an illustrative embodiment. The method 2200 begins and proceeds to operation 2202, where the environment analytics system 118 monitors and captures location updates for a plurality of users navigating the indoor environment 102. From operation 2202, the method 2200 proceeds to operation 2204, where the environment analytics system 118 presents the location updates on a display in accordance with color codes associated with each granular area of the indoor environment 102. From operation 2204, the method 2000 proceeds to operation 2206, where the method 2200 ends. The method 2200 shows the location updates on the layout), (Id., ¶ 152, The UI application can be executed by the processor 2604 to aid a user in interacting with at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other data associated with the indoor environment, the environment analytics system 118, the network 116, and/or the other systems 122. The UI application can be executed by the processor 2604 to aid a user in answering/initiating calls, entering/deleting other data, entering and setting user IDs and passwords for device access, configuring settings, manipulating address book content and/or settings, multimode interaction, interacting with other applications 2610, and otherwise facilitating user interaction with the operating system 2608, the applications 2610, and/or other types or instances of data 2612 that can be stored at the mobile device 2600. The data 2612 can include, for example, at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other applications or program modules);
PNG
media_image2.png
431
579
media_image2.png
Greyscale
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the animated heatmap elements of Ozkan in the analogous art of determining indoor location of devices using sensors for the same reasons as stated for claim 1.
Regarding Claim 12, the combination of Lobo and Ozkan discloses …The method of claim 11…
Lobo further discloses …further comprising, iterating to the deriving when the current time elapses by a preconfigured amount of time to update the scores and the heatmap based on updated store data for the store (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates (discloses heatmap data structure)), (Id., ¶ 38, At 318, predictive modeling is conducted using the retail shrinkage data 122, the external data 132, and the issuance of an alert 142. Conducting predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. At 320, an alert is issued if the predictive modeling determines that a high shrinkage risk situation is likely to occur. (discloses updates based on store data) In some cases, if the predictive modeling determines that a high shrinkage risk situation is likely to occur, the sensor control system 110 can be caused to alter a setting 106 of at least one of the plurality of sensors 104).
Regarding Claim 13, the combination of Lobo and Ozkan discloses …The method of claim 11…
Lobo further discloses … further comprising, providing feature identifiers for the features and the scores to applications associated with the store systems (Id., ¶ 23, Sensor control system 110 is communicatively coupled to sensors 104 located and arranged in a retail environment 102. In general, the sensor control system 110 is configured to control one or more settings 106 for each of the sensors 104. Further, the sensor control system 110 receives real-time sensor data 108 from the sensors 104 as well. Retail environment 102 can include any store or physical, defined retail space.), (Id., ¶ 27, Analytics engine 140 is communicatively coupled with first shrinkage database 120. Accordingly, analytics engine 140 can access retail shrinkage data 122, including information regarding items 124 at high risk for shrinkage and times 126 at high risk for shrinkage activity. Analytics engine 140 is also communicatively coupled with the second shrinkage database 130, and accordingly, has access to external data 132. Similarly, analytics engine 140 is also communicatively coupled with the sensor control system 110, such that analytics engine 140 can receive real-time sensor data 108 from sensors 104. Communications may be wired, wireless, and may rely on a variety of known techniques, protocols, or communication technology).
Regarding Claim 14, the combination of Lobo and Ozkan discloses …The method of claim 13…
Lobo further discloses …further comprising, training the MLM based on actual shrink events identified in the store data (Lobo, ¶ 25, First shrinkage database 120 includes retail shrinkage data 122 for the retail environment 102. The retail shrinkage data 122 can include one or more items 124 at high risk for shrinkage. These may be items that have a history of being stolen frequently, are particularly valuable, or are known to be related to frequent shrinkage-related issues. Items that are small, easy to conceal, or difficult to track could also be deemed items 124 at high risk for shrinkage. The retail shrinkage data 122 can include one or more times 126 at high risk for shrinkage activity. These times 126 can include times of day when shrinkage is most common, times of the week common for shrinkage, times of the year common for shrinkage, or times of expected shrinkage related to holidays and local activities. Further, certain items can be correlated to certain times to identify high shrinkage risk. In some embodiments, at least one item 124 at high risk or at least one time 126 at high risk for shrinkage is part of the retail shrinkage data 122), (Id., ¶ 42, Initially at 501, an individual is detected and an image is captured at 503. Next, a procedure is used to identify one or more features of the individual at 505 using the analytics engine. The system is queried as to whether a match for the feature(s) was found at 507. If no match is found at 507, the device control engine tunes the sensors at 509 and tunes the camera at 511. If a match for the feature(s) is found at 507, and a maximum number of attempts for searching the matched feature(s) is not exceeded at 513, the central cloud 540 searches criminal reports and one or more known high risk individual databases at 515. The shrink risk is assessed at 517, and if a high risk of shrink is found, alerts 519 are given. Further, cameras are rotated appropriately to the individual detected at 521 and video is captured of the individual and his or her movements at 523. Further if alerts are given at 519, clients are notified at 525 and criminal reports and known high risk individual databases are updated at 527. This information is fed into the machine learning engine 550 for training at 529 and predictions are made by the machine learning engine 550 at 531).
Regarding Claim 15, the combination of Lobo and Ozkan discloses …The method of claim 11…
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose …wherein deriving further include retaining predictive values provided by other MLMs from the store systems within the store data as first features.
However, Ozkan discloses …wherein deriving further include retaining predictive values provided by other MLMs from the store systems within the store data as first features (Ozkan, ¶ 134, Turning now to FIG. 19, a method 1900 for calculating and displaying heat maps on a store layout will be described, according to an illustrative embodiment. In some embodiments, one or more of the operations of the method 1900 are performed by the environment analytics system 118 in response to input from a user. In some other embodiments, one or more of operations of the method 1900 are automated by the environment analytics system 118 using machine learning algorithms and/or other techniques (discloses other MLMs from the store systems) programmed into the analytics application 128), (Id., ¶ 138, From operation 1914, the method 1900 proceeds to operation 1916, where the environment analytics system 118 enters a first while loop. In particular, while X.sub.2<X.sub.Max, X.sub.1 is set equal to X.sub.2 and X.sub.2 is set equal to X.sub.2+G and operations 1912 and 1914 are repeated. From operation 1916, the operation 1900 proceeds to operation 1918, where the environment analytics system 118 enters a second while loop. In particular, while Y.sub.2<Y.sub.Max, Y.sub.1 is set equal to Y.sub.2 and Y.sub.2 is set equal to Y.sub.2+G and operations 1912, 1914 and 1916 are repeated. Turning now to FIG. 20, a graph 2000 illustrating a store layout on a two-dimensional coordinate system and a granularity (“G”) 2002 used to calculate a heat map for a store layout is illustrated, according to an illustrative embodiment).
Regarding Claim 16, the combination of Lobo and Ozkan discloses …The method of claim 15…
Lobo further discloses …wherein deriving further includes obtaining external data associated with dates of transactions and a physical location of the store as second features (Lobo, ¶ 5, In an embodiment, a retail shrinkage activity prediction and identification system includes: a sensor control system, a first shrinkage database, a second shrinkage database, an analytics engine, and a machine learning engine. The sensor control system is communicatively coupled with a plurality of sensors arranged in a retail environment. The sensor control system is configured to control a setting of each of the plurality of sensors. The first shrinkage database includes retail shrinkage data for at least the retail environment. The retail shrinkage data includes at least one item at high risk for shrinkage or at least one time at high risk for shrinkage activity. The second shrinkage database includes external data related to shrinkage in a geographic area of the retail environment (discloses external location data)), (Id., ¶ 6, The analytics engine is communicatively coupled with: the first shrinkage database to access the retail shrinkage data, the second shrinkage database to access the external data, and the sensor control system to receive real-time sensor data from the plurality of sensors. The analytics engine is configured to compare the real-time sensor data with the external data to identify a high shrinkage risk situation. If a high shrinkage risk situation is identified, the analytics engine will: issue an alert, cause the sensor control system to alter the setting of at least one of the plurality of sensors, and update at least one of the first shrinkage database or the second shrinkage database. The machine learning engine is communicatively coupled with the first shrinkage database, the second shrinkage database, and the analytics engine to use the retail shrinkage data, the external data, and the issuance of an alert to conduct predictive modeling and cause the analytics engine to issue an alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur), (Id., ¶ 32, Sensors 104 are configured to obtain real-time sensor data 108 by sensing a characteristic of the environment, a product or structure (e.g., a shelf, modular, door, cart, basket, etc.), individuals, activities or movements of individuals or groups, timing or length of events, dates, times, or other potentially relevant data to shrink in a retail environment 102).
Regarding Claim 17, the combination of Lobo and Ozkan discloses …The method of claim 15…
Lobo further discloses …wherein deriving further includes identifying third features as transactions and terminal types associated with the transactions (Lobo, ¶ 31, Referring to FIG. 2, a retail environment 102 is depicted in which a plurality of sensors 104 is located in various locations throughout the retail environment 102. Sensors 104 may be located in fixed locations or may be mobile. In some embodiments, sensors 104 can solely or primarily include security cameras located throughout a store. Alternatively or additionally, sensors 104 can include infrared sensors, optical sensors, temperature sensors, pressure sensors, or other non-intrusive sensors. Sensors 104 may be mounted at or proximate to a store entrance or exit 103, in locations above or in various aisles 105 or shelves 107, on the walls 109, on the ceiling or fixtures, in the floor, on carts or baskets, or at any other suitable site of a retail environment 102. In some embodiments, a sensor 104 is mounted proximate to each lane of a point-of-sale (POS) system 111 or checkout area 113. (discloses terminal type feature data) In some cases, sensors 104 such as pressure or temperature sensors can be mounted or arranged in the floor or walking surfaces of retail environment 102. Sensors 104 that are mobile may be mounted on or in an unmanned aerial vehicle, drone, robot, ceiling structure or substructure, or floor structure permitting movement).
Regarding Claim 18, the combination of Lobo and Ozkan discloses …The method of claim 15…
Lobo further discloses … wherein deriving further includes identifying additional features associated with identities of cashiers for the transactions, identities of customers for the transactions, and payment methods used in the transactions (Id., ¶ 18, Some embodiments can turn cameras toward more susceptible items or areas at certain times. Some embodiments can update an internal shrink database to better track and proactively identify high risk items and areas of a store. Some embodiments can also utilize machine learning to identify an individual in a retail store by capturing an image of the individual and searching external or public reports and databases of high-risk individuals or groups, such as those shared among retailers or business associations in particular industries and/or geographic areas or available from government or law enforcement agencies), (Id., ¶ 28, In general, analytics engine 140 compares or processes real-time sensor data 108 with external data 132 to identify high shrinkage risk situations. In some embodiments, comparing or processing the real-time sensor data 108 with the external data 132 to identify a high shrinkage risk situation comprises matching an image in the real-time sensor data 108 with an image in the external data 132. For example, an image of an individual may be compared to images in a police or local database of individuals having a history of shoplifting or related criminal offenses. In other embodiments, comparing the real-time sensor data 108 with the external data 132 comprises identifying and correlating sensor data events and/or patterns in sensor data 108 with data in or extracted from external data 132. This identifying and correlating, as well as the comparing or processing more generally, can include a single point or factor analysis (e.g., does sensor data element A match, correlate with and/or lead to external data element Z), multi-point/factor analysis (e.g., do sensor data elements A, B and C, respectively, match, correlate with, and/or lead to external data elements Z, Y and X; does the aggregate of sensor data elements A, B and C match, correlate with, and/or lead to the aggregate of external data elements Z, Y, and X; etc.), or a combination of single and multiple point/factor analysis), (Id., ¶ 32, Sensors 104 are configured to obtain real-time sensor data 108 by sensing a characteristic of the environment, a product or structure (e.g., a shelf, modular, door, cart, basket, etc.), individuals, activities or movements of individuals or groups, timing or length of events, dates, times, or other potentially relevant data to shrink in a retail environment 102).
Regarding Claim 19, Lobo discloses …A system, comprising: a cloud server comprising at least one processor and a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium comprising executable instructions, wherein the executable instructions, when executed by the at least one processor cause the at least one processor to perform operations comprising: training a machine learning model (MLM) on features derived from store data obtained from store systems and data stores of a store to produce as output a time series of sets of scores over a predefined period of time that predicts at each interval a corresponding set of scores indicating how likely each feature or combinations of the features is likely or not likely to be associated with shrink during the corresponding interval, wherein the MLM provides the time series for a business day at predetermined intervals of time and continuously updates further time intervals for the business day based on current data being received from the store (Lobo, ¶ 41, Referring to FIG. 5, a flowchart of an example of a method 500 of predicting or identifying retail shrinkage is shown. In general, the flowchart of the method 500 is divided into columns corresponding to various parts of the system in which actions occur. In general, these include: camera and motion and sensor detection 510; device control engine 520; shrink analysis engine 530; central cloud 540; and machine learning engine 550. In certain embodiments, the camera and motion and sensor detection 510 can correspond to the sensors 104 described earlier in this disclosure. Likewise, device control engine 520 can be understood to generally correspond to server control system 110, shrink analysis engine 530 and central cloud 540 can together generally correspond to the analytics engine 140, and machine learning engine 550 corresponds to machine learning engine 150), (Id., ¶ 43, system 100 and/or its components or systems can include computing devices, microprocessors, modules and other computer or computing devices, which can be any programmable device that accepts digital data as input, is configured to process the input according to instructions or algorithms, and provides results as outputs. In an embodiment, computing and other such devices discussed herein can be, comprise, contain or be coupled to a central processing unit (CPU) configured to carry out the instructions of a computer program. Computing and other such devices discussed herein are therefore configured to perform basic arithmetical, logical, and input/output operations), (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur), (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates), (Id., ¶ 25, First shrinkage database 120 includes retail shrinkage data 122 for the retail environment 102. The retail shrinkage data 122 can include one or more items 124 at high risk for shrinkage. These may be items that have a history of being stolen frequently, are particularly valuable, or are known to be related to frequent shrinkage-related issues. Items that are small, easy to conceal, or difficult to track could also be deemed items 124 at high risk for shrinkage. The retail shrinkage data 122 can include one or more times 126 at high risk for shrinkage activity. These times 126 can include times of day when shrinkage is most common, times of the week common for shrinkage, times of the year common for shrinkage, or times of expected shrinkage related to holidays and local activities. Further, certain items can be correlated to certain times to identify high shrinkage risk. In some embodiments, at least one item 124 at high risk or at least one time 126 at high risk for shrinkage is part of the retail shrinkage data 122);
PNG
media_image3.png
310
451
media_image3.png
Greyscale
receiving real-time store data at a current interval of time from the store systems and the data stores; deriving current features from the real-time store data; providing the current features as input to the MLM (Id., ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features (discloses customer data and employee data). Real-time sensor data 108 can also include data related to movements of individuals or groups, (discloses security data) congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, (discloses transaction data) length of trip data, or other potentially relevant tracked information), (Id., ¶ 31, Referring to FIG. 2, a retail environment 102 is depicted in which a plurality of sensors 104 is located in various locations throughout the retail environment 102. Sensors 104 may be located in fixed locations or may be mobile. In some embodiments, sensors 104 can solely or primarily include security cameras located throughout a store. Alternatively or additionally, sensors 104 can include infrared sensors, optical sensors, temperature sensors, pressure sensors, or other non-intrusive sensors), (Id., ¶ 39, FIG. 4 shows a diagram of an example of a store network 400 in which a retail shrinkage activity prediction and identification system can be implemented. In FIG. 4, the store network generally includes a store 410 having a main retail environment 412 and a back office 414. The main retail environment 412 includes a rotating system 416 having a plurality of related POS systems 418, cameras 420, and connectivity devices 422. Also connected to the main retail environment are maintenance resources 424);
receiving as output current sets of current scores for current features and current combinations of the features, wherein a last set of the current scores ends at a close of business hours for the store (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur), (Id., ¶ 25, First shrinkage database 120 includes retail shrinkage data 122 for the retail environment 102. The retail shrinkage data 122 can include one or more items 124 at high risk for shrinkage. These may be items that have a history of being stolen frequently, are particularly valuable, or are known to be related to frequent shrinkage-related issues. Items that are small, easy to conceal, or difficult to track could also be deemed items 124 at high risk for shrinkage. The retail shrinkage data 122 can include one or more times 126 at high risk for shrinkage activity. These times 126 can include times of day when shrinkage is most common, times of the week common for shrinkage, times of the year common for shrinkage, or times of expected shrinkage related to holidays and local activities. Further, certain items can be correlated to certain times to identify high shrinkage risk. In some embodiments, at least one item 124 at high risk or at least one time 126 at high risk for shrinkage is part of the retail shrinkage data 122);
obtaining a planogram for a physical layout of the store with resource identifiers for resources of the store identified within the physical layout; generating a heatmap data structure with labels or images for the resource identifiers superimposed on top of the physical layout (Id., ¶ 17, Embodiments relate to systems and methods for prediction and identification of retail shrinkage activity. Prediction and identification of factors presenting high shrink risk enables surveillance resources and monitoring to be beneficially allocated to mitigate these risks. Some embodiments can utilize a machine learning engine to identify items that are susceptible to shrinkage and the times of day they are likely to be stolen. Certain embodiments can identify items that are more susceptible to shrinkage by shoplifters. Embodiments can use information from past shoplifting incidents to identify high risk items. Embodiments can receive information from external or public data systems regarding high risk items in the past. Embodiments can include an in-store shrink analysis system. Embodiments can provide reports, including heat maps, to store managers and associates);
and obtaining external data associated with dates of transactions and a physical location of the store as second features (Id., ¶ 20, Retail stores or environments in which these shrink mitigation systems and methods can be used include virtually any retail outlet, including a physical, brick-and-mortar storefront; or some other setting or location via which a customer may purchase or obtain products. Though only the case of a single retail environment is generally discussed in examples used herein, in many cases, the systems and methods can include a plurality of retail environments. For example, data from one or a plurality of retail environments can be aggregated, analyzed and applied to one or a plurality of other retail environments. In some embodiments, data from one or a plurality of retail environments can be aggregated, analyzed and/or applied in conjunction with data related to other shopping behaviors, patterns or other factors), (Id., ¶ 24, Sensors 104 can include a plurality of sensors. The plurality of sensors 104 can include any of a surveillance camera, an optical sensor, a motion detection sensor, a temperature sensor, an infrared sensor, a microphone, or a pressure sensor, for example. Settings 106 of each of the sensors 104 can include an activation, a direction, an angle, a zoom level, a location or a sensing area, for example. Real-time sensor data 108 can include image data, such as an image of clothing or facial features. Real-time sensor data 108 can also include data related to movements of individuals or groups, congregating of individuals, temperature profile data, infrared data, sound recording data, pressure data, time of purchase data, length of trip data, or other potentially relevant tracked information).
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose …providing a function with the heatmap data structure to animate the certain resources, the certain combinations, and the corresponding indicia within the physical layout over time until the close of the business hours for the store; and rendering the heatmap data structure within an interactive interface on a device operated by a store user for the user to visually identify areas within the store, the certain resources, and combinations of the certain resources that are associated with high risk of shrink throughout the business hours of the store; and assigning a unique color based on the current scores to icons or graphical images representing certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores.
However, Ozkan discloses …providing a function with the heatmap data structure to animate the certain resources, the certain combinations, and the corresponding indicia within the physical layout over time until the close of the business hours for the store; and rendering the heatmap data structure within an interactive interface on a device operated by a store user for the user to visually identify areas within the store, the certain resources, and combinations of the certain resources that are associated with high risk of shrink throughout the business hours of the store (Ozkan, ¶ 34, Fig. 22 is a flow diagram illustrating a method for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map, (discloses animated heatmap) according to an illustrative embodiment), (Id., ¶ 49, The environment analytics system 118 can perform operations to interact with the environment database 120, and more particularly, environment data 132, customer data 134, product data 136, promotion data 138, beacon data 140, heat map data 142 and/or calibration data 144 stored within the environment database 120. The environment analytics system 118 can save data to the environment database 120, retrieve data from the environment database 120, delete data from the environment database 120, edit data and saved edited data to the environment database 120, and manipulate data stored within the environment database 120. The environment analytics system 118 can provide data retrieved from the environment database 120 to the user device 110 and/or the other systems 122 via the network 116), (Id., ¶ 58, The heat map data 142 can include data associated with customer heat maps. Customer heat maps are used herein to identify hot spots, dead areas and bottlenecks of customer traffic within the indoor environment 102. Customer heat maps can aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results. Companies can use heat maps to visualize the impact of changes to the indoor environment in terms of customer flows, sold items, average sales values, and the like. The heat map data 142 can include previously recorded customer coordinate during a given time interval and/or real-time customer coordinate information. (discloses timeline function) It should be understood that the heat map data 142 can include any combination of the aforementioned data and other data associated with heat maps that is not specified herein), (Id., ¶ 137, From operation 1912, the method 1900 proceeds to operation 1914, where the environment analytics system 118 assigns a heat map color (discloses heat map color coding) for each area based upon the number of unique user records. For example, if the number of unique users is 0, the heat map color may be white; for 1-10 unique users, the heat map color may be light yellow; for 11-50 users, the heat map color may be dark yellow; for 51-100 unique users, the heat map color may be orange; for 101-200 unique users, the heat map color may be brown; and for greater than 200 unique users, the heat map color may be red. These colors are provided as examples only and the actual number of unique users for each heat map color may be specified differently for different implementations of the concepts and technologies disclosed herein. An example heat map is shown in grayscale in FIG. 21);
and assigning a unique color based on the current scores to icons or graphical images representing certain resources and certain combinations of the certain resources that are associated with the current features and current combinations based on corresponding current scores Id., ¶ 8, the environment analytics system also can apply a coordinate system to the layout of the environment. The coordinate system can include the absolute reference point. The environment analytics system also can determine a minimum coordinate pair, a maximum coordinate pair, a granularity, and a time interval. The environment analytics system also can set a first coordinate equal to a first minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a second coordinate equal to a second minimum coordinate of the minimum coordinate pair. The environment analytics system also can set a third coordinate equal to a sum of the first minimum coordinate and the granularity, and can set a fourth coordinate equal to a sum of the second minimum coordinate and the granularity. The environment analytics system also can query the environment database for a number of unique user location records with a first location coordinate between the first coordinate and the third coordinate, a second location coordinate between the second coordinate and the fourth coordinate, and a timestamp within the time interval. The environment analytics system also can determine heat map color codes for a plurality of different numbers of unique user location records. The environment analytics system also can generate a heat map that includes a plurality of areas representing at least a portion of the heat map color codes), (Id., ¶ 137, From operation 1912, the method 1900 proceeds to operation 1914, where the environment analytics system 118 assigns a heat map color for each area based upon the number of unique user records. For example, if the number of unique users is 0, the heat map color may be white; for 1-10 unique users, the heat map color may be light yellow; for 11-50 users, the heat map color may be dark yellow; for 51-100 unique users, the heat map color may be orange; for 101-200 unique users, the heat map color may be brown; and for greater than 200 unique users, the heat map color may be red. (discloses assigning colors to an image based on current scores) These colors are provided as examples only and the actual number of unique users for each heat map color may be specified differently for different implementations of the concepts and technologies disclosed herein. An example heat map is shown in grayscale in FIG. 21), (Id., ¶ 140, Turning now to FIG. 22, a method 2200 for determining location updates for a plurality of users and presenting the location updates on a store layout as a heat map will be described, according to an illustrative embodiment. The method 2200 begins and proceeds to operation 2202, where the environment analytics system 118 monitors and captures location updates for a plurality of users navigating the indoor environment 102. From operation 2202, the method 2200 proceeds to operation 2204, where the environment analytics system 118 presents the location updates on a display in accordance with color codes associated with each granular area of the indoor environment 102. From operation 2204, the method 2000 proceeds to operation 2206, where the method 2200 ends. The method 2200 shows the location updates on the layout), (Id., ¶ 152, The UI application can be executed by the processor 2604 to aid a user in interacting with at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other data associated with the indoor environment, the environment analytics system 118, the network 116, and/or the other systems 122. The UI application can be executed by the processor 2604 to aid a user in answering/initiating calls, entering/deleting other data, entering and setting user IDs and passwords for device access, configuring settings, manipulating address book content and/or settings, multimode interaction, interacting with other applications 2610, and otherwise facilitating user interaction with the operating system 2608, the applications 2610, and/or other types or instances of data 2612 that can be stored at the mobile device 2600. The data 2612 can include, for example, at least a portion of the environment data 132, at least a portion of the customer data 134, at least a portion of the product data 136, at least a portion of the promotion data 138, at least a portion of the beacon data 140, at least a portion of the heat map data 142, and/or other applications or program modules).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the animated heatmap elements of Ozkan in the analogous art of determining indoor location of devices using sensors for the same reasons as stated for claim 1.
Regarding Claim 20, the combination of Lobo and Ozkan discloses …The system of claim 19…
Lobo further discloses …wherein the executable instructions when executed by the at least one processor further cause the at least one processor to perform additional operations, comprising: providing feature identifiers for the current features and the corresponding current scores for the current features and the current combinations of features to one or more workflows processed by the store systems… (Id., ¶ 30, Machine learning engine 150 is communicatively coupled with the first shrinkage database 120, the second shrinkage database 130, and the analytics engine 140. Machine learning engine 150 is able to receive and utilize retail shrinkage data 122, external data 132, and the issuance of an alert 142 to conduct predictive modeling. Predictive modeling can include identifying trends in any combination of the retail shrinkage data 122, the external data 132, the issuance of an alert 142, and the real-time sensor data 108 that relate to high shrinkage risk situations. In one example, a thousand items could be present at one location, where 30% of the items are deemed high risk, 20% of the items are deemed medium risk, and all the rest of the items are considered low risk. Based on historical data, the times and locations when shrink happened are known. Accordingly, it is possible to use a time series prediction or another machine learning algorithm to find any trends present. This enables prediction of when high risk situations will happen for those items which are high or medium risk. Further, the machine learning engine 150 can cause the analytics engine 140 to issue a further alert if the predictive modeling determines that a high shrinkage risk situation is likely to occur. In some embodiments, the analytics engine 140 is configured to cause the sensor control system 110 to alter the setting 106 of at least one of the plurality of sensors 104 if the predictive modeling determines that a high shrinkage risk situation is likely to occur), (Id., ¶ 33, Referring to FIG. 3, a flowchart of a method 300 of predicting or identifying retail shrinkage activity is shown. In general, the method 300 includes first accessing retail shrinkage data including at least one item at high risk for shrinkage or at least one time at high risk for shrinkage activity in a retail environment 102, at 302. An item at high risk for shrinkage could be certain high-priced electronics or pharmaceuticals that frequently are stolen, for example. A time at high risk for shrinkage could be between 4-5 pm in some locations, for example.
While suggested in at least Fig. 2 and related text, Lobo does not explicitly disclose …through an application programming interface.
However, Ozkan discloses …through an application programming interface (Ozkan, ¶ 151, The UI application can interface with the operating system 2608 to facilitate user interaction with functionality and/or data stored at the mobile device 2600 and/or stored elsewhere, such as in the environment database 120. In some embodiments, the operating system 2608 can include a member of the SYMBIAN OS family of operating systems from SYMBIAN LIMITED, a member of the WINDOWS MOBILE OS and/or WINDOWS PHONE OS families of operating systems from MICROSOFT CORPORATION, a member of the PALM WEBOS family of operating systems from HEWLETT PACKARD CORPORATION, a member of the BLACKBERRY OS family of operating systems from RESEARCH IN MOTION LIMITED, a member of the IOS family of operating systems from APPLE INC., a member of the ANDROID OS family of operating systems from GOOGLE INC., and/or other operating systems. These operating systems are merely illustrative of some contemplated operating systems that may be used in accordance with various embodiments of the concepts and technologies described herein and therefore should not be construed as being limiting in any way).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo to include the application interface elements of Ozkan in the analogous art of determining indoor location of devices using sensors for the same reasons as stated for claim 1.
Claims 7-9 are rejected under 35 U.S.C. 103 as being unpatentable over Lobo in view of Ozkan and Wong and in further view of Shivashankar et al., U.S. Publication No. 2019/0108561 [hereinafter Shivashankar].
Regarding Claim 7, the combination of Lobo, Ozkan and Wong discloses …The method of claim 6…
While suggested in at least Fig. 2 and related text of Lobo, the combination of Lobo and Ozkan does not explicitly disclose …wherein training further includes obtaining weather data associated with a date of each transaction for a location of the store and generating a weather feature for the corresponding transaction based on the weather data.
However, Shivashankar discloses …wherein training further includes obtaining weather data associated with a date of each transaction for a location of the store and generating a weather feature for the corresponding transaction based on the weather data (Shivashankar, ¶ 54, The purchase intent determination and assistance management system (PIDAMS) computes and refines the dwell time threshold periodically based on the iterative statistical inputs. The PIDAMS extracts daily climatic conditions, for example, minimum temperature, maximum temperature, weather conditions such as raining, snowing, thunder storm, etc., at the location of the retail store. The shopper demographics, for example, gender and age also affect the configuration of the dwell time threshold. If the shoppers include only children, or only adults, or adults accompanied by family, for example, by children, the PIDAMS configures the dwell time thresholds differently. In an example, if the identified anonymous shopper is single, the identified anonymous shopper may spend less time in a configured region of interest, and hence, the PIDAMS configures a short dwell time threshold at the configured region of interest. In another example, if the identified anonymous shopper is accompanied by a child, the identified anonymous shopper may spend more time in a configured region of interest of a section, for example, a stationary section or a toy section of the retail store. The PIDAMS, in this example, configures a longer dwell time threshold at the configured region of interest of the stationary section or the toy section of the retail store), (Id., ¶ 95, FIGS. 7A-7B exemplarily illustrate graphical representations showing dwell time distributions of anonymous shoppers identified in configured regions of interest that are in view of sensors, for example, camera-1 and camera-2 in a retail store, indicating performance of iterative statistical models in determining dwell time thresholds. The iterative statistical models learn from the received metadata, the weather conditions, and the shift roster data. The purchase intent determination and assistance management system (PIDAMS) computes the dwell time of the identified anonymous shoppers as disclosed in the detailed description of FIG. 2. The iterative statistical models compute the dwell time thresholds as disclosed in the detailed description of FIGS. 5-6. The average dwell time of the identified anonymous shoppers in a configured region of interest in view of camera-1 increases as the day proceeds as exemplarily illustrated in FIG. 7A, and the average dwell time of the identified anonymous shoppers in a configured region of interest in view of camera-2 increases and then decreases as the day proceeds as exemplarily illustrated in FIG. 7B. In an embodiment, the trained iterative statistical models, using the received metadata, the weather conditions, and the shift roster data, compute the dwell time threshold for a region of interest to be less than the dwell time threshold computed by the trained iterative statistical models using only the received metadata. The iterative statistical models calculate the dwell time thresholds in a batch mode at the end of a current day using historical data of dwell time obtained until the end of working hours of the current day of the retail store).
It would have been obvious to a person of ordinary skill in the art before the effective filing date to have modified the store shrinkage scoring elements of Lobo, the heatmap interface elements of Ozkan and the random forest elements of Wong to include the weather data elements of Shivashankar in the analogous art of purchase intent determination.
The motivation for doing so would have been to improve an ability of “iteratively ranking anonymous shoppers, converting the anonymous shoppers into potential buyers, and improving shopper experience in a retail store, thereby increasing revenue for the retail store” (Shivashankar, ¶ 8), wherein such improvements would benefit Ozkan’s method which seeks to improve customer service and experience by “preventing fraud in cases where the physical presence of a payment card cannot be ascertained (e.g., online transactions referred to as “card-not-present”) or for commercial transactions where high-value transactions may be routine and where it may be difficult to classify patterns of behaviour as “unexpected”” [Wong, ¶ 25], wherein such improvements would beneifit Ozkan’s method which seeks to “aid, wherein such improvements would further benefit Ozkan’s method which seeks to “aid companies to optimize store performance, to improve customer service and to improve marketing and promotion results” [Ozkan, ¶ 58], wherein such benefits would benefit Lobo’s method which seeks to improve an “ability to deter shrinkage from stores in a proactive way would be extremely useful to retailers while also improving retail customer experience” [Shivashankar, ¶ 8; Wong, ¶ 25; Ozkan, ¶ 58; Lobo, ¶ 4].
Regarding Claim 8, the combination of Lobo, Ozkan, Wong and Shivashankar discloses …The method of claim 7…
Lobo further discloses …wherein training further includes identifying at least one risk score provided in the security data for one or more of a corresponding transaction, a customer, a terminal, and a cashier, and generating at least one additional feature for the at least one risk score (Id., ¶ 28, In general, analytics engine 140 compares or processes real-time sensor data 108 with external data 132 to identify high shrinkage risk situations. In some embodiments, comparing or processing the real-time sensor data 108 with the external data 132 to identify a high shrinkage risk situation comprises matching an image in the real-time sensor data 108 with an image in the external data 132. For example, an image of an individual may be compared to images in a police or local database of individuals having a history of shoplifting or related criminal offenses. (discloses risk score for a customer) In other embodiments, comparing the real-time sensor data 108 with the external data 132 comprises identifying and correlating sensor data events and/or patterns in sensor data 108 with data in or extracted from external data 132. This identifying and correlating, as well as the comparing or processing more generally, can include a single point or factor analysis (e.g., does sensor data element A match, correlate with and/or lead to external data element Z), multi-point/factor analysis (e.g., do sensor data elements A, B and C, respectively, match, correlate with, and/or lead to external data elements Z, Y and X; does the aggregate of sensor data elements A, B and C match, correlate with, and/or lead to the aggregate of external data elements Z, Y, and X; etc.), or a combination of single and multiple point/factor analysis), (Id., ¶ 42, Initially at 501, an individual is detected and an image is captured at 503. Next, a procedure is used to identify one or more features of the individual at 505 using the analytics engine. The system is queried as to whether a match for the feature(s) was found at 507. If no match is found at 507, the device control engine tunes the sensors at 509 and tunes the camera at 511. If a match for the feature(s) is found at 507, and a maximum number of attempts for searching the matched feature(s) is not exceeded at 513, the central cloud 540 searches criminal reports and one or more known high risk individual databases at 515. The shrink risk is assessed at 517, and if a high risk of shrink is found, alerts 519 are given. Further, cameras are rotated appropriately to the individual detected at 521 and video is captured of the individual and his or her movements at 523. Further if alerts are given at 519, clients are notified at 525 and criminal reports and known high risk individual databases are updated at 527. This information is fed into the machine learning engine 550 for training at 529 and predictions are made by the machine learning engine 550 at 531).
Regarding Claim 9, the combination of Lobo, Ozkan, Wong and Shivashankar discloses …The method of claim 8…
Lobo further discloses …wherein training further includes generating second features for transactions based on 1) calendar dates relative to weekdays, weekend, and known holidays, 2) basket items for the corresponding transaction, 3) checkout channel for the corresponding transaction, 4) a cashier identifier for the corresponding transaction, 5) a loyalty member of the store or a non-loyalty member associated with a customer for the corresponding transaction; and 6) a method of payment for the corresponding transaction (Lobo, ¶ 42, The shrink risk is assessed at 517, and if a high risk of shrink is found, alerts 519 are given. Further, cameras are rotated appropriately to the individual detected at 521 and video is captured of the individual and his or her movements at 523. Further if alerts are given at 519, clients are notified at 525 and criminal reports and known high risk individual databases are updated at 527. This information is fed into the machine learning engine 550 for training at 529 and predictions are made by the machine learning engine 550 at 531), (Id., ¶ 25, The retail shrinkage data 122 can include one or more times 126 at high risk for shrinkage activity. These times 126 can include times of day when shrinkage is most common, times of the week common for shrinkage, times of the year common for shrinkage, or times of expected shrinkage related to holidays (discloses holiday features) and local activities. Further, certain items can be correlated to certain times to identify high shrinkage risk. In some embodiments, at least one item 124 at high risk or at least one time 126 at high risk for shrinkage is part of the retail shrinkage data 122), (Id., Sensors 104 are configured to obtain real-time sensor data 108 by sensing a characteristic of the environment, a product or structure (e.g., a shelf, modular, door, cart, basket, etc.), (discloses basket features) individuals, activities or movements of individuals or groups, timing or length of events, dates, times, or other potentially relevant data to shrink in a retail environment 102), (Id., ¶ 31, Sensors 104 may be mounted at or proximate to a store entrance or exit 103, in locations above or in various aisles 105 or shelves 107, on the walls 109, on the ceiling or fixtures, in the floor, on carts or baskets, or at any other suitable site of a retail environment 102. In some embodiments, a sensor 104 is mounted proximate to each lane of a point-of-sale (POS) system 111 or checkout area 113. (discloses checkout channel and payment method) In some cases, sensors 104 such as pressure or temperature sensors can be mounted or arranged in the floor or walking surfaces of retail environment 102. Sensors 104 that are mobile may be mounted on or in an unmanned aerial vehicle, drone, robot, ceiling structure or substructure, or floor structure permitting movement), (Id., ¶ 19, References to “shrinkage” or “shrink,” as used throughout this disclosure, are intended to refer generally to loss of inventory that can be attributed to factors such as theft, shoplifting, administrative errors, fraud, and cashier errors (discloses cashier features) that benefit the purchaser), (Id., ¶ 42, an individual is detected and an image is captured at 503. Next, a procedure is used to identify one or more features of the individual at 505 using the analytics engine. The system is queried as to whether a match for the feature(s) was found at 507. If no match is found at 507, the device control engine tunes the sensors at 509 and tunes the camera at 511. If a match for the feature(s) is found at 507, and a maximum number of attempts for searching the matched feature(s) is not exceeded at 513, the central cloud 540 searches criminal reports and one or more known high risk individual databases at 515. The shrink risk is assessed at 517, and if a high risk of shrink is found, alerts 519 are given. Further, cameras are rotated appropriately to the individual detected at 521 and video is captured of the individual and his or her movements at 523. Further if alerts are given at 519, clients are notified at 525 and criminal reports and known high risk individual databases are updated at 527. This information is fed into the machine learning engine 550 for training at 529 and predictions are made by the machine learning engine 550 at 531).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Fisher et al., U.S. Patent No. 10,133,933 discloses item put and take detection using image recognition.
Buibas et al., U.S. Publication No. 2021/0049772 discloses an automated store that tracks shoppers who exit a vehicle.
Shaw, U.S. Publication No. 2015/0058049 discloses systems and methods for identifying and analyzing a customer queue.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICHOLAS D BOLEN whose telephone number is (408)918-7631. The examiner can normally be reached Monday - Friday 8:00 AM - 5:00 PM PST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Patty Munson can be reached on (571) 270-5396. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/NICHOLAS D BOLEN/Examiner, Art Unit 3624
/PATRICIA H MUNSON/Supervisory Patent Examiner, Art Unit 3624